-
Notifications
You must be signed in to change notification settings - Fork 52
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
remove respawn in fetch_bringup/launch/include/teleop.launch.xml #44
Conversation
@knorth55 could you tell us why? |
I'm not sure why we would want to disable the respawn=true on this for all Fetch robots. If @knorth55 can give us a compelling reason, then maybe we can make it an optional thing that some users could disable if they want to. |
other nodes related to teleop are running as |
@moriarty kindly ping. could you give me some feedback? |
@knorth55 your above explanation still doesn't describe why this change should be made. What undesired effect is the respawn parameter having? Having respawn be an arg that defaults to true, would probably be our preference. I changed the PR to merge into melodic-devel, so you will need to rebase or cherry-pick your commit accordingly. I don't think it makes sense to update indigo-devel at this point. |
@erelson We want to set same value for |
Also, we are using fetch in Indigo, so I want to backport this PR to indigo branch, too. |
It sounds like you want the teleop node to not respawn if it exits on startup due to the existence of the same node already. Since the teleop node is launched via /etc/ros/$distro/robot.launch, for your usecase would it make sense to comment out the line in that file? (This file is intended to be user-modified, and doesn't get overwritten by updates) Also, note that there will not be any further releases of fetch ROS debians for ROS Indigo. |
My use case is like this.
In this case I want to set I mean there is no reason to set Indigo is EOL, so I need to switch to use this package from source :( |
The two nodes which are re-spawned are the topic_tools mux, and the tuck arm script. This is odd. I would have expected us to respawn the core required components... But I can't think of why we would respawn only the tuck_arm_script unless it's known to crash frequently, or the topic mux. |
The topic mux should not respawn. In fact it should probably be listed as required. Tuck arm should also not respawn. It doesn't crash |
@cjds yes I think if we wanted to enforce some nodes were always running the better way would be to use required. If we were to change these all to respawn=true that would be a bigger change. I think we can accept his PR. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Seems Reasonable
I'm fine with merging at this point. It looks like the respawns were not added in response to issues with the nodes crashing, so the consensus is they're safe to remove.
This may be a mis-statement, as it doesn't make sense to me? robot.launch is initially a copy of fetch.launch, with (minimal) modifications such as pointing to a calibrated URDF. |
Yes, you are right. |
We want to remove
respawn
param for some teleop nodes